1

The Co-Evolution of Technology and Methods of Standard Setting: the case of the mobile phone industry

Jeffrey L. Funk

Professor, Hitotsubashi University, Institute of Innovation Research 2-1 Naka, Kunitachi, Tokyo 186-8603 Japan

Telephone: 81-42-580-8430, Facsimile: 81-42-580-8410, funk@iir.hit-u.ac.jp Abstract

This paper applies the concept of co-evolution to technology, institutions, and industry structure in the mobile phone industry with a focus on technology and the institution of standard setting. The paper finds that technological change has caused the method of standard setting to come full circle. Technological change initially supported the move from quasi-vertical integration to an open standard setting process in both wireline and mobile phone systems where changes in the method of standard setting lagged the changes in technology. Growth in those markets that initially implemented an open standard setting process in mobile phone systems encouraged other countries to adopt similar types of standard setting where government agencies and firms were the mechanisms for this transmission of open standard setting methods. However, technological change in the form of the mobile Internet has caused quasi-vertical integration to return in the form of service providers determining the mobile Internet standards and the specifications for the phones that support their mobile Internet services. Again changes in the method of standard setting are lagging changes in technology and a new set of firms is transmitting these methods of standard setting to the rest of the world.

Keywords: technology, evolution, transmission, standards, global,

JEL Classifications: L15, L22, L43, O30, O32, O33

2

1. Introduction

Evolutionary theorists have long been concerned with the interaction between

technology, institutions, and industry structure, and why some countries implement

institutions and industry structure that support economic growth faster than other

countries (Schumpeter, 1942; Rosenberg, 1976; Nelson and Winter, 1982). Nelson

(2002) emphasizes the role of routines and the role they play in both technological

change and institutions where it is the evolution of these routines that determines

economic growth.

Strategic management theorists use different terms to represent similar concepts.

Although they also use the terms technology and industry structure, they often use the

terms process and organizational structure instead of routine and institution respectively

(Mintzberg, 1978). For example, Henderson and Clark’s (1990) contingency model

emphasizes the need for different processes and forms of organizational structures to

handle different forms of technological change such as integral versus modular problem

solving. And because most industries undergo a gradual evolution from integral to

modular problem solving over time, particularly as a dominant design (Abernathy and

Utterback, 1978; Anderson and Tushman, 1990; Utterback, 1994) emerges, firms tend to

develop processes and organizational structures that support modular problem solving.

These processes and structures can act as “core rigidities” for innovations that occur less

frequently (Leonard-Barton, 1992; Christensen, 1997) and that require a different form

of problem solving such as integral problem solving (Clark and Fujimoto, 1991;

Chesbrough, 2003).

Whether we use the terms of strategic management or evolutionary theory,

co-evolutionary models can help us understand the interaction between technological

3

change, institutions, and industry structure (Lewin et al, 1999; Nelson, 1995; Ziman,

1999). Recent work on co-evolution has called for analysis of both directions of

causality (Pelikan, 2003), stricter definitions of co-evolution, and their application to

institutions as opposed to merely firms (Murmann, 2003).

This paper applies the concept of co-evolution to technology, institutions, and to a

lesser extent industry structure in the mobile phone industry where standard setting is

considered the key institution. Although there is a substantial literature on standard

setting in many industries (Grindley, 1995; David, 1987; Shapiro and Varian, 1999;

David and Steinmueller, 1994), the network effects associated with these interface

standards (Katz and Shapiro, 1986; Farrell and Saloner, 1985), the impact of standard

setting on competition between firms (Katz and Shapiro, 1985; Katz and Shapiro, 1994)

and between countries (Funk and Methe, 2001), and the role of standard setting as a

quasi-legal institution (Antonelli, 1994), little of this research has looked at standard

setting as an evolutionary process or the effect of technological change on the

appropriate method of standard setting. This is in spite of the fact that standard setting

and the numbers of institutions that support standard setting have dramatically increased

over the last fifty years in a wide range of industries.

This paper uses the mobile phone industry to demonstrate this co-evolution of

technology and standard setting because there have been several technological changes

during the last 25 years in this industry and standards have played a critical role in each

of these changes. The unit of analysis is the method of standard setting, the method of

standard setting can be considered a set of routines, and Durham’s (1991) model of

evolutionary change is used to characterize the evolution of standard setting including

the methods and processes of transmission of these standard setting methods. As subsets

4

of the main interaction between technological change and the method of standard setting,

this paper also considers their interactions with competition and patent policy, which

can be considered outputs from institutions, and with industry structure. This paper uses

a narrow definition of industry structure in order to differentiate it from standard setting.

It defines industry structure in terms of whether a service provider does the majority of

research and development work or whether other firms such as manufacturers do.

With respect to technological change and the method of standard setting, this paper

finds that changes in the method of standard setting have lagged changes in technology

where the changes in technology have enabled new forms of problem solving. Of

greatest interest, the evolution in problem solving and standard setting has been a

circular rather than a uni-directional process. In the first three generations of mobile

phone technology, there was a gradual move from integral to modular problem solving

and the emergence of standard setting processes that support this modular problem

solving. The most recent technological change, however, the mobile Internet, requires

integral problem solving and this has caused the emergence of a method of standard

setting that is similar to what existed in the wireline industry in the late 1970s.

2. Previous research

Beginning with Penrose (1959), management scholars and later strategic

management theorists have emphasized that organizational structure should match the

underlying technology and they have created many types of contingency models to

support this notion (Mintzberg, 1978). For example, Henderson and Clark’s (1990)

contingency model emphasizes the need for different forms of problem solving in

modular and architectural innovations. Architectural innovations require different forms

5

of processes and organizational structure than do modular ones where scholars use

terms such as “integral problem solving,” to describe the type of problem solving that is

needed to handle architectural innovations (Clark and Fujimoto, 1991; Iansiti, 1997;

Chesbrough, 2003; Steinmueller, 2003).

Other management scholars have focused on the ability of firms to change. Some

argue that organizations possess some capacity to adapt to changing environments

(Nelson and Winter, 1982; Nadler and Tushman, 1997; Teece, Pisano, Shuen, 2000).

Others argue that firms tend to develop processes that support for example incremental

and modular innovations and these processes act as a “core rigidity” for architectural

and other innovations that occur less frequently and that require other integral problem

solving (Leonard-Barton, 1992; Christensen, 1997; Chesbrough, 2003).

Evolutionary theorists also consider the ability of firms and other institutions to

change. Nelson and Winter’s (1982) groundbreaking research on evolutionary theories

of technological change emphasized routines. Some of Nelson’s (1995) subsequent

research has emphasized co-evolution and the interaction between social and physical

technologies where the concept of routines is used to describe these technologies. For

example, the social routines of economies of scale and professional management only

became productive at the end of the 19th century because they were needed by the new

physical routines of mass production (Nelson, 2002; Chandler, 1962). Similarly, the

commercialization of synthetic dyestuffs required new physical routines of synthesis

and new social routines of organizing chemists in the laboratory and training them in

universities (Nelson, 2002; Murman, 2003).

The concept of social and physical routines can also be applied to standard setting.

The growth in the broadcasting, computer, electronics, telecommunication, video and

6

audio recording, and related industries in the 20th century has increased the importance

of physical routines such as the definition of standards and the importance of social

routines such as those of standard setting institutions. These standard setting institutions

range from government bodies such as the U.S. National Institute of Standards and

Technology to international organizations such as the International Telecommunications

Union, professional organizations such as the Institute of Electric and Electronic

Engineers, and multi-firm coalitions that are created to support specific standards for

example for audio or video recording (Farrell and Saloner, 1988; David and

Steinmueller, 1994; Shapiro and Varian, 1999). The range of standard setting institutions

reflects the large variety of social routines that can be defined for them.

The large variety of these social routines and the standard setting institutions they

are found in suggest that the concept of managerial contingency models can also be

applied to institutions such as standard setting ones. Just as firms must match the

processes and organizational structure to the method of problem solving, firms either by

themselves or in alliances with other firms and governments must also match the social

routines of standard setting to the method of problem solving where this paper focuses

on the difference between modular and integral problem solving. And drawing parallels

with the organizational change literature, this paper uses the concept of co-evolution to

consider how these methods of standard setting evolve along with the technology. It

defines the evolution of standard setting in terms of Durham’s (1991) five aspects of

evolutionary change: 1.) units of transmission; 2.) sources of variation; 3.) mechanisms

of transmission; 4.) processes of transformation; and 5.) sources of isolation.

7

3. Methodology

This paper primarily relies on published research on wireline (Brock, 1981; Brock,

1994: David and Steinmueller, 1994; Fransman, 2002), first generation mobile phone

(King and West, 2002; Lyytinen and Fomin, 2002; Lehenkari and Miettinen, 2002),

second generation mobile phone (Haug, 2002; Bekkers et al, 2002; Funk, 2003), overall

mobile phone (Garrard, 1998; Funk and Methe, 2001; Funk, 2002), and mobile Internet

(Fransman, 2002; Natsuno, 2003; Sigurdson, 2001) standard setting. Due to the high

rate of recent change in the mobile Internet, we also supplemented the published sources

with consulting reports and interviews with the largest service providers, technology

providers, and phone manufacturers in the world.

Between 2001 and 2005, multiple interviews were conducted with multiple

managers of NTT DoCoMo, KDDI, Vodafone, T-Mobile, Hutchison Telecom, Sprint,

Cingular Wireless, Nokia, Ericsson, and Qualcomm Japan, and Qualcomm Korea.

Because these interviews revealed that the failure of the global standard setting process

for mobile Internet standards, WAP (Wireless Automation Protocol), had forced

individual service providers to define their own mobile Internet specifications and have

custom phones made according to these specifications (Economist, 2005; Reinhardt and

Ihlwan, 2005), we focused our recent interviews on the largest service providers in the

world and also investigated the history of their services via press releases on their sites.

4. Results: A brief history of standard setting in the mobile phone industry

Following Durham (1991), Table 1 defines the general categories of an evolutionary

process for standard setting in the mobile phone industry. The unit of transmission is the

method of standard setting and the sources of variation are different countries with

8

different methods of standard setting. In the process of transformation, governments and

firms have adopted new forms of standard setting for a variety of reasons including to

promote domestic growth and domestic equipment producers. Government agencies,

multi-national firms, and organizations act as the mechanisms of transmission. Sources

of isolation include language, culture, and government policies.

Table 2 summarizes technological change in the wireline and mobile phone

industries and its impact on problem solving, the dominant form of standard setting,

competition policy, and industry structure. Within wireline, it differentiates between pre-

and post-deregulation and within mobile phones, it differentiates between three

generations of air-interface standards and the mobile Internet. Table 3 provides more

details on the interaction between technological change and the evolution of standard

setting where the table differentiates between the setting of the air-interface standards

and the determination of phone specifications. Table 4 lists the most widely used

air-interface standards.

Partly driven by changes in the wireline industry, there was an evolution towards

more open standard setting in the first three generations of mobile phone air-interface

standards. The U.S. and Scandinavia moved the fastest followed by the rest of Europe

and Japan (See Figure 1 and Table 3). As graphically represented in Figure 2, the faster

growth in those markets that adopted more open forms of standard setting and

competition encouraged other countries to also introduce open standards and

competition. The diffusion of open standard setting and competition also caused

research and development to move from away from service providers and towards

manufacturers and other firms (change in industry structure), which strengthened the

open standard setting processes and patent protection. It has only been in the latest

9

generation of technological change (the mobile Internet) that the dominant form of

standard setting has come full circle and returned to its original form of quasi-vertical

integration due to the need for integral problem solving.

Place Tables 1-4 and Figures 1 and 2 about here

4.1 Wireline

Until the 1980s, telecommunication industries in most advanced countries were in a

state of quasi-vertical integration. Regulated monopolies argued that the possibilities of

harm to the network from the attachment of unauthorized equipment required them to

be responsible for determining all the specifications in a wireline system from switching

equipment to phones (Brock, 1981; Brock, 1994; David and Steinmueller, 1994). In this

integral problem solving the regulated monopolies ordered equipment often from single

vendors where the same phones, usually black ones, were used by almost everyone in a

single country. It was only in less advanced countries that service providers purchased

turnkey systems from vendors and thus there was greater competition between

equipment vendors than there was in advanced countries (Fransman, 2002).

There was a slow move towards modular problem in the second half of the 20th

century that accelerated in the 1970s and 1980s as firms and governments began to

realize that service providers did not need to be responsible for defining all the

specifications in a telephone system (Fransman, 2002). From the technological side, the

move to modular problem solving was driven by the use of electronic switches, which

reduced the “coupling” between the phones and switches (Talley, 1982; David and

Steinmueller, 1994; Steinmuller, 2003) partly because the electronic switches handled

10

the speech and signaling paths separately. Although electronic switching was first

introduced in the 1960s (Huurdeman, 2003), the implementation of modular design also

required regulatory changes that came about in the U.S. through a number of court cases

in the 1950s, 1960s, and 1970s where the lack of harm to the network from each change

gradually convinced the Federal Communications Commission (FCC) and other parties

that a fully open interface in which phones were attached to network via standard plugs

as opposed to hard wiring would be beneficial to consumers. It became possible for

consumers in 1978 to purchase phones from manufacturers other than AT&T’s

manufacturing subsidiary Western Electric where these manufacturers were free to

determine the specifications and designs for phones in any way they liked as long as the

phones conformed to the open interface (Brock, 1981, 1994). The definition of this open

interface also supported the Justice Department’s breakup of AT&T that enabled

competition between service providers to increase in the early 1980s. Other large

advanced countries such as Great Britain, France, and Germany followed with similar

changes in the mid-to late 1980s (Fransman, 2002).

4.2 Analog Mobile

The different rates at which countries defined an open interfaces between phones

and networks and deregulated their wireline industries was the primary source of

variation in the method of standard setting in analog mobile systems. Just as the U.S.

and Scandinavia ones were the first countries to define open interfaces between the

phone and network in the wireline industries, they were also the first countries to do so

in analog mobile systems (Funk and Methe, 2001, Lyytinen and Fomin, 2002). In the

U.S., the FCC was the primary mechanism of transmission while the overall driver of

11

change was the Justice Department. The FCC had learned from its experience in the

wireline industry that modular problem solving was both possible and advantageous to

consumers and based on proposals from manufacturers it defined an open interface for

an “Advanced Mobile Phone System” (AMPS) and published specifications for the

interface (Lyytinen and Fomin, 2002). The definition of this open interface coincided

with the breakup of AT&T, which was a result of a ten-year court battle with the Justice

Department. As it defined the open interface, the FCC also divided the U.S. into more

than 700 regions and awarded licenses to two firms in each of them in a complex drawn

out process (Garrard, 1998; King and West, 2002).

In Scandinavia, it was both competition with non-Scandinavian countries and

cooperation with other Scandinavian countries that encouraged the service providers

and regulatory agencies to implement open standard setting processes and modular

problem solving (Fransman, 2002). National service providers began defining interfaces

between the wireline network and phones and other key building blocks within the

network and ordering equipment from multiple vendors in the 1960s. These national

service providers also cooperated on the creation of a pre-cellular1 and later on an

analog cellular mobile phone system that was called the Nordic Mobile Telephone

(NMT) system (Lehenkari and Miettinen, 2002; Lyytinen and Fomin, 2002). Using the

terminology of evolutionary processes, regulatory agencies and service providers were

the means by which the open method of standard setting was transmitted (mechanism of

transmission) from wireline to pre-cellular and cellular systems (Fransman, 2002).

Although competition between mobile phone service providers within Scandinavia was

not introduced until digital services (see below) were started in the early 1990s, users

1 Pre-cellular systems used a single transmitter to cover a large area.

12

were allowed to purchase phones and thus there was competition between phone

manufacturers (Hulten and Molleryd, 1995; Lehenkari and Miettinen, 2002).

Scandinavian countries began introducing analog mobile phone services based on

NMT in 1981 and the U.S. based on AMPS in 1983. While other advanced countries

such as Japan, Germany, France, and Italy also introduced their own unique systems,

there was little growth in their systems due to the lack of open standards, the control of

these standards by the national service providers, and a lack of competition between

either service providers or between manufacturers all of which reflected a continuation

of quasi-vertical integration by the national service providers. By the end of 1986, more

than 80% of the world’s subscribers used phones that were based on NMT, AMPS, or a

variant of AMPS and most of these subscribers were either in North America,

Scandinavia, or Great Britain (Garrard, 1998; Funk, 2002).

4.3. Digital Mobile

The greater success of the Scandinavian and U.S. mobile phone markets encouraged

other countries to introduce open standard setting processes and modular problem

solving for digital systems (See Figure 1 and Table 2). Germany and France, which had

previously tried to convince Great Britain to develop an alternative to the U.S. and

Scandinavian analog systems, had turned their attention to digital systems by 1984

(Funk, 2002) and the creation of an open system that would later become a global

standard. Using a committee called Group Special Mobile that had been created in 1982

by the Conference Posts and Telecommunications (CEPT), Germany and France created

an alliance of Western European governments, service providers, and later

manufacturers that developed the GSM standard, whose name was later to become

13

Global System Mobile. By early 1987, 15 countries had signed a memorandum of

understanding (MOU) to introduce digital systems that were based on the GSM

standard (Meurling and Jeans, 1994; Garrard, 1998).

Scandinavian government agencies and to a lesser extent service providers and

manufacturers such as Ericsson and Nokia acted as the mechanisms of “transmission”

of this open approach to standard setting in Western Europe. These government

agencies and firms convinced Germany, France, and other European countries to adopt

Scandinavia’s open form of standard setting in which service providers initially played

the key role (Haug, 2002). The service providers in CEPT evaluated a number of

proposals from manufacturers and chose a basic design in 1987 (Meurling and Jeans,

1994; Garrard, 1998; Haug, 2002) where a growing desire for economic integration

within Western Europe and competition with the U.S. and Japan also helped convince

European countries to cooperate with each other (Funk and Methe, 2001; Funk, 2002).

It was only with the creation of ETSI (European Telecommunications Standards

Institute) that manufacturers began to become the dominant players in standard setting

for GSM and thus the mobile phone industry. The standard setting work for GSM was

transferred from the CEPT to ETSI in January 1988 where a key decision by ETSI was

to use subscriber information module (SIM) cards, which enabled a phone to be used

with any GSM service provider merely by exchanging the SIM cards (Hawkins, 1993;

Garrard, 1998) thus further strengthening modular problem solving

The transfer of standard setting to ETSI and the use of SIM cards reflected the

increasing power of the manufacturers and also contributed to it (See Figure 2). During

the 1980s, manufacturers had been steadily increasing their development spending in

the wireline and mobile industries while service providers had been reducing their

14

spending (Fransman, 2002). The manufacturers achieved a major victory in GSM when

the service providers revoked their claim for a common IPR (Intellectual Property

Rights) policy in the late 1980s (Bekkers et al, 2002). Stronger patent protection further

strengthened the role of manufacturers in both development and standard setting in

GSM and helped them become global agents of diffusion for the GSM standard (Funk

and Methe, 2001).

Countries that did not adopt GSM as a national standard include the U.S., Japan, and

Korea. The U.S. extended its concept of competition from that between service

providers and manufacturers to that between standards (See Figure 1) and it auctioned

the frequency spectrum as opposed to awarding the spectrum on the basis of a “beauty

contest” (Garrard, 1998; Scanlon, 2001). Following the auctions, services were started

in the U.S. that were based on several standards including GSM, TDMA, and CDMA

(Scanlan, 2001) where CDMA was based on Quallcom’s technology. Although there are

many versions of Quallcom’s CDMA technology, this paper calls Quallcom’s 2nd

generation technology cdmaOne and its 3rd generation technology cdma2000.

Korea and Japan also did not adopt GSM as a national standard and thus both

countries and in particular Japan continued their isolation in an evolutionary sense.

Korea adopted cdmaOne as its digital standard in the mid-1990s while Japan allowed its

major service provider, NTT DoCoMo to define the national standards for analog and

initially digital systems (Funk, 2002; Funk, 2003).

The common aspect of Japan and Korea’s isolation was that they did not use GSM

and SIM cards. This slowed the introduction of global roaming, the complete

introduction of modular problem solving between phones and services, and the

movement of development spending from service providers to manufacturers. Service

15

providers continued to define the specifications for phones (Funk, 2003), which can be

seen as a partial continuation of quasi-vertical integration (See Table 3). Unlike the West

where open standards and SIM cards had completely destroyed quasi-vertical

integration, it has continued to partially exist in Japan and Korea. As shall be described

in section 4.5, the Japanese and Korean service providers have used their ability to

define phone specifications and standards for mobile Internet services to ensure the

consistent display of content across different phones (See Table 3).

4.4. Third Generation

The success of GSM further strengthened the policies of openness, modular problem

solving, competition, manufacturer development, and patent protection in the third

generation standard setting process. By the end of the 1990s, the GSM Alliance had

become a global organization with a global standard setting process that was integrated

with the processes of the International Telephone Union (ITU). Qualcomm’s desire to

promote its version of CDMA called cdmaOne and to compete with the GSM standard

caused it to form the CDMA Development Group, which also became integrated with

the ITU and now plays the same role as the GSM alliance does (Funk, 2002). Here it

was both manufacturers and service providers that were the mechanisms for the

transmission of open standard setting from the GSM Alliance to the CDMA

Development Group.

Governments also increased the amount of competition by increasing the number of

licenses and awarding them via auctions as opposed to “beauty contests.” By the late

1990s, many countries had awarded third generation licenses via these auctions (See

Table 2) some of which commanded more than 50 Billion Euros (BBC, 2000). By the

16

early 2000s the number of countries announcing their plans to liberalize the choice of

air-interface standard (including Europe) had also increased (Scanlan, 2001; Economist,

2000).

Outside of Japan, manufacturers did most of the development work and patents

played an even larger role in third generation than in second generation standard setting.

Qualcomm’s large number of patents in CDMA and its decision not to offer mobile

phone infrastructure or handsets (it initially entered and withdrew) further increased the

importance of patents where there were major conflicts between Qualcomm and the

traditional manufacturers like Ericsson and Nokia in the late 1990s and early 2000s2.

The success of GSM also finally had an impact on standard setting in Japan thus

breaking down the walls of its isolation (in the evolutionary sense) in the late 1990s.

Differences in language and culture and the fact that the Japanese government allowed

NTT DoCoMo to control the analog and digital phone standard setting processes were

some of the reasons for Japan’s previous isolation. Domestic and foreign manufacturers

and service providers worked to end this isolation and this sense they can be seen as one

of the mechanisms of transmission of an open approach in third generation standard

setting in Japan. Domestic manufacturers felt that Japan’s adoption of proprietary

systems in analog and digital systems had isolated them from the global market while

other Japanese service providers complained about NTT DoCoMo’s domination of

domestic standard setting. Foreign manufacturers such as Motorola, Nokia, and

Ericsson and service providers such as Vodafone also pushed for more openness in order

to facilitate their participation in the Japanese market. These complaints caused the

Japanese government to demand in the mid-1990s that NTT DoCoMo either adopt or

2 For example, see (WSJ, 1998).

17

create a global standard for third generation systems (Funk and Methe, 2001; Funk,

2002) and to approve the use of a non-Japanese second generation digital system

(cdmaOne), which Japan’s second largest service provider (KDDI) adopted.

As they had done in second generation standard setting, Scandinavian

organizations, albeit here it was Ericsson and Nokia, were an additional mechanism of

transmission of a new approach to standard setting for Japan. NTT DoCoMo had chosen

to create a global standard with its W-CDMA (Wide-Band CDMA) technology and

Ericsson and Nokia agreed to support W-CDMA in exchange for having the GSM

network interface included in the global standard. They did this partly because there was

a growing consensus in the late 1990s that Qualcomm’s cdmaOne technology was

technologically superior to GSM and might be chosen as a third generation standard by

Europe. This was a major threat to Nokia and Ericsson since they were not supplying

any cdmaOne infrastructure (Funk and Methe, 2001; Funk, 2002). Ericsson and Nokia

used this concession concerning the network interface to convince European service

providers to adopt W-CDMA in ETSI in January 1998. Subsequently most service

providers that use GSM have introduced W-CDMA while the initial adopters of

cdmaOne have introduced cdma2000 and its improved versions.

4.5. Mobile Internet

The success of open air-interface standards and the setting of them by manufacturers

in the first three generations of mobile phone systems initially strengthened the policies

of openness and manufacturer development in standard setting for the mobile Internet.

Nokia, Motorola, and Ericsson acted as the mechanisms of transmission in an

evolutionary sense in that they created the WAP (Wireless Application Protocol) Forum

18

in June 1997. There were almost 100 members by early 1999 and more than 500 by

mid-2001 (Sigurdson, 2001).

However, in spite of the emphasis on openness, the manufacturers were unable to

agree on standards in the mobile Internet. The lack of agreement on standards led to an

inconsistent display of menus and content across different phones and it required users

to configure the services themselves; both of these problems substantially reduced

interest in the services (Sigurdson, 2001; Fransman, 2002). Furthermore, the ability to

use GSM phones in every country, which is a major advantage in voice applications,

became a disadvantage as every GSM service in the world experienced the same

problems with phones.

On the other hand, Japanese and Korean service providers continued to define

phone specifications and maintain a partial system of quasi-vertical integration (See

Table 3). In Japan, NTT DoCoMo defined its own mobile Internet standards, had

manufacturers develop phones that conformed to these standards, and introduced

i-mode in February 1999 (Fransman, 2002; Natsuno, 2003). The success of i-mode

caused other Japanese and Korean service providers to introduce similar services in

1999 and 2000. Japan’s second and third largest service (KDDI and J-Phone) and

Korean’s three largest service providers (SK Telecom, KT Freetel, and LG) also define

their own standards and have manufacturers supply them with phones that are

customized for their mobile Internet services.

The difference in performance between Japan, Korea, and the rest of the world

reflects changes in the appropriate method of problem solving and thus the need for a

different method of standard setting (See Table 2). While the Western approaches to

standard setting worked well with modular problem solving, the mobile Internet

19

requires integral problem solving. As opposed to the single air-interface standard that

was handled with an open standard setting process, there are multiple interfaces in the

mobile Internet where each interface involves a different application and the importance

of an application must be recognized before a standard can be set for the application. It

is well recognized that individual firms can identify these market opportunities better

than standard setting committees (Shapiro and Varian, 1999) and this was certainly the

case in the mobile Internet. Initially the WAP Forum did not even try to define standards

for entertainment applications such as ringing tones, screen savers, and games, which

have become the killer applications in the mobile Internet.

Furthermore, consistency between the multiple interface standards is needed and

this consistency must be maintained as changes in technology require updates to the

multiple standards. For example, each type of entertainment content requires a specific

method of formatting ringing tones, screen savers, and games, these formats must be

consistent with other phone software such as browsers, Java virtual machines,

mail/messaging clients, and packet and micro-payment systems3, and these formats and

other software are updated as improvements in displays, processors, and other chips are

implemented. Maintaining the consistency between multiple interface standards requires

integral problem solving by service providers.

Although the process of change is still underway, the greater success of the

Japanese and Korean service providers has caused Western service providers to also

define phone specifications, order custom phones, and thus reintroduce a partial form of

quasi-vertical integration. In an evolutionary sense, the mechanisms of transmission are

3 In the micro-payment systems, service providers bill the users for purchases of content and give a percentage of these fees to the content providers.

20

service providers such as NTT DoCoMo and Vodafone and Japanese and Korean phone

manufacturers. NTT DoCoMo has licensed its i-mode service to more than 10

non-Japanese service providers. Vodafone has used its investment in Japan’s third

largest service provider, J-Phone (called Vodafone Japan until Softbank acquired it in

2006) to understand the necessary specifications for the mobile Internet and introduce a

global service called Vodaphone Live!. Korean and Japanese phone manufacturers were

the first ones to offer custom phones to Vodafone and other Western service providers

such as Sprint PCS, Hutchison Telecom, and Verizon Wireless where Nokia and

Motorola did not supply these custom phones in large volumes until 2004 (Economist,

2005; Reinhardt and Ihlwan, 2005).

This change in standard setting has put the West significantly far behind Japan and

Korea in the mobile Internet (Natsuno, 2003). Vodafone did not introduce its mobile

Internet service, Vodafone Live! until more than 3/12 years after NTT DoCoMo

introduced i-mode and other Western service providers have been much slower to set

specifications and obtain custom phones. Furthermore, several key aspects of the NTT

DoCoMo’s i-mode service such as Internet mail on all phones have not been introduced

outside of Japan. Instead, SMS is still the dominant form of messaging service outside

of Japan and the prices for SMS are 5-15 times the price in Europe and the U.S. as the

prices for mobile mail are in Japan4.

Nevertheless, a convergence in standard setting in the mobile Internet is occurring

4 NTT DoCoMo charges 1 Yen (0.008 Euros at 125 Yen per Euro) to receive a short Internet mail message (http://www.nttdocomo.co.jp/english/p_s/charges/mova/f/imode.html) versus 0.136 Euros to receive an SMS in Europe (Credit Suisse, 2004). Other service providers do not charge users to receive Internet mail in some plans (e.g., Vodafone Japan). http://www.vodafone.jp/english/live/mail/skymail.html

21

and standard setting in the mobile Internet now partially resembles that of the wireline

industry in the late 1970s. Just as the service providers in the largest countries were able

to determine the specifications and obtain custom equipment for their wireline systems

more than smaller countries were able to do in the late 1970s (Fransman, 2002), the

largest mobile phone service providers can now obtain custom phones much easier than

smaller ones can, which place the smaller service providers at a large disadvantage. On

the other hand, as mobile Internet services evolve and mature, it is likely that an open

interface or a dominant design (Abernathy and Utterback, 1978; Anderson and Tushman,

1990; Utterback, 1994) will emerge between the phones and networks that will enable a

return to modular problem solving.

The major difference between the wireline industry in the late 1970s and the

mobile Internet in 2005 is that the large service providers are now global and not

national providers and they order their phones from global and not national

manufacturers. This is most evident in Europe where a few global players offer most of

the mobile phone services. The competition between mobile Internet services is now

between global service providers who are trying to convince global manufacturers to

make custom phones for them. And unlike the late 1970s, national governments have far

less power to influence this competition. Time and further research will tell us about the

merits and demerits of this case of “globalization.”

5. Discussion

The goal of this paper was to explore the co-evolution of technology, institutions,

and industry structure in the mobile phone industry with a focus on technological

change and the institution/method of standard setting. This paper finds that the methods

22

of standard setting have lagged changes in the method of problem solving that were

brought about by technological change. Initially, different countries with different

methods of standard setting were the sources of variation in this co-evolutionary process.

Standard setting processes that led to growth in specific countries were adopted by other

countries in the next generation of technology where government agencies,

multi-national firms, and organizations acted as the mechanisms of transmission of

these standard setting processes.

The most interesting aspect of this co-evolution of technology and standard setting

is its circular nature. During the 1960s, 1970s, and 1980s some firms and governments

began to recognize that it was possible to create an open interface between phones and

networks in wireline and later mobile phone systems and thus use modular in place of

integral problem solving. The move from first to second and third generation mobile

phone systems strengthened this modular problem solving and the open standard setting

processes that supported it. However, the mobile Internet has changed the method of

problem solving from modular to integral and this has caused the method of standard

setting to return to a situation that is partly similar to one that existed for wireline in the

late 1970s. The mobile Internet now requires integral problem solving as opposed to the

modular problem solving needed for setting air-interface standards. This has caused the

quasi-vertical integration of Japanese and Korean service providers to work better than

the “open” standard setting processes of WAP where the existence of the WAP Forum

and the support by manufacturers for it act as a “core rigidity” (Leonard-Barton, 1992).

We can also think of this change in standard setting as a change in physical and

social routines. Air-interface standards can be thought of as a collection of physical

routines for physically transmitting signals between a single interface that connects

23

mobile phones and base stations. On the other hand, mobile Internet standards can be

thought of as physical routines that involve multiple interfaces between phones, servers

(i.e., content providers), and services (i.e., service providers) where each interface

involves a different application and the important entertainment applications were not

initially recognized. Both the identification of these applications and the setting of

standards for them require a different set of social routines than are needed for the

determination of the air-interface standards.

The social routines of standard setting have been transmitted by firms and

government agencies. In the second and third generation systems, Scandinavian

government agencies and firms (Ericsson and Nokia) were major transmitters of these

social routines from their successful first generation standard setting processes and this

may be one reason why Ericsson and Nokia still remain the leading suppliers of

infrastructure and phones respectively. It is too early to forecast how recent changes in

standard setting in the mobile Internet will impact on these firms. While it appears that

the change in problem solving and thus the appropriate method of standard setting favor

service providers such as NTT DoCoMo and Korean manufacturers like Samsung, it is

possible that the integral problem solving is merely a temporary situation until a

dominant design (Anderson and Tushman, 1990) emerges as it did for example in the

personal computer (PC) industry where vertically integrated solutions gave way to

modular ones following the introduction of the IBM PC (Langlois, 1993).

In addition to the co-evolution of technology and standard setting, this paper has

also touched on the co-evolution of competition and patent policy and industry structure.

Open standard setting facilitated the introduction of competition in phones and services

and led to growth in mobile phone markets first in Scandinavia and the U.S. This

24

growth led to further support for open standard setting and competition in these and

other mobile phone markets (See Figure 2). By the early 1990s, most countries had

introduced competition in mobile phone services and these services were based on open

standards and open standard setting processes. Increased openness in the standard

setting process also reflected a move in development activities from service providers to

manufacturers where increasing returns to development (Klepper, 1996) and stronger

patent protection probably accelerated this move. The greater openness in the standard

setting processes used in Western markets caused this to happen faster in the West than

in Japan and Korea. The slower move to open standards in Japan and to a lesser extent

Korea caused NTT DoCoMo and to a lesser extent other Japanese and Korean service

providers to retain their development capability and exert more control over the phone

specifications.

Future research should consider the co-evolution of technology and standard setting

in other industries. The increasing number of industries that require standards, the

growing number of standard setting institutions, and the large variety in these

institutions suggest that this is a fruitful area of research. Industries such as computers,

music, and broadcasting have experienced large amounts of change in both technology

and the method of standard setting and it is likely that there has been an interaction

between them. Such research will likely shed more light on the process of co-evolution.

25

6. References

Abernathy W and Utterback J (1978) Patterns of innovation in technology. Technology

Review 80: 40-47.

Anderson P and Tushman M (1990) Technological discontinuities and dominant

designs: A cyclical model of technological change. Administrative Science Quarterly

35: 604-633.

Antonelli C (1994) Localized technological change and the evolution of standards as

economic institutions. Information economics and policy 6: 195-216.

BBC (British Broadcasting Service) (2000) Europe's mobile phone rollercoaster.

October 23, http://news.bbc.co.uk/1/hi/business/986428.stm.

Bekkers R, Verspagen B, Smits J (2002) Intellectual property rights and standardization:

the case of GSM. Telecommunications Policy 26: 171-188.

Brock G (1981) The Telecommunications Industry: The Dynamics of Market Structure.

Cambridge: Harvard University Press.

Brock G (1994) Telecommunication Policy for the Information Age: From Monopoly to

Competition, Cambridge, MA: Harvard University Press.

Chandler A (1962) Strategy and structure: chapters in the history of the industrial

enterprise. Cambridge: MIT Press.

Chesbrough H (2003) Towards a Dynamics of Modularity: A Cyclical Model of

Technical Advance, in Prencipe A, Davies A, Hobday M (eds) The Business of

Systems Integration, NY: Oxford University Press.

Christensen C (1997) The innovator’s dilemma, Boston: Harvard Business School

Press.

Clark K and Fujimoto T (1991) Product Development Performance: Strategy,

26

Organization, and Management in the World Auto Industry. Boston: Harvard

Business School Press.

David P and Steinmueller E (1987) Some new standards for the economics of

standardization in the information age, In: Dasgupta P. and Stoneman P (eds)

Economic policy and technological performance, ch. 8. Cambridge: Cambridge

University Press.

David, P. (1994) Economics of compatibility standards and competition in

telecommunication networks, Information Economics and Policy 6: 217-241.

Durham W (1991) Coevolution: Genes, Culture and Human Diversity. Stanford

University Press.

Economist (2000) The price is right, July 27.

Economist (2005) The giant in the palm of your hand, February 12, pp. 59-61.

Farrell, J. and G. Saloner (1985) Standardization, Compatibility and Innovation. RAND

Journal of Economics 16(1) 70 – 83.

Farrell, J. and G. Saloner (1988) Coordination through Committees and Markets. RAND

Journal of Economics 19(2): 235 – 252.

Fransman M (2002) Telecoms in the Internet Age: From Boom to Bust To...? Oxford:

Oxford University Press.

Funk J and Methe D (2001) Market- and Committee-based Mechanisms in the Creation

and Diffusion of Global Industry Standards: The Case of Mobile Communication.

Research Policy 30: 589-610.

Funk J (2002) Global Competition Between and Within Standards: the case of mobile

phones. London: Palgrave.

Funk J (2003) Standards, dominant designs and preferential acquisition of

27

complementary assets through slight information advantages. Research Policy:

32(8): 1325-1341.

Garrard G (1998) Cellular Communications: Worldwide Market Development. Boston:

Artech House.

Grindley P (1995) Standards strategy and policy: Cases and stories. Oxford University

Press.

Haug T (2002) A commentary on standardization practices: lessons from the NMT and

GSM mobile telephone standards histories. Telecommunications Policy 26(3-4):

101-107.

Hawkins R (1993) Changing expectations: voluntary standards and the regulation of

European telecommunications, Communications and Strategies 11: 53-85.

Henderson R and Clark K (1990) Architectural innovation: The reconfiguration of

existing product technologies and the failure of established Firms. Administrative

Science Quarterly 35: 9-30.

Hulten S and Molleryd B (1995) Mobile Telecommunications in Sweden. In: Schenk K,

Muller J, Schnoring T (eds) Mobile Telecommunications: Emerging European

Markets. Boston and London: Artech House.

Huurdeman, A. (2003) The Worldwide History of Telecommunications. NY: John Wiley

& Sons.

Iansiti M (1997) Technology Integration: Making Critical Choices in a Dynamic World.

Boston: Harvard Business School Press.

Katz M and Shapiro C. (1985) Network Externalities, Competition, and Compatibility.

American Economic Review 75(3): 424-440.

Katz M and Shapiro C (1986) Technology Adoption in the Presence of Network

28

Externalities. The Journal of Political Economy 94(4): 822-841.

Katz M and Shapiro C (1994) Systems Competition and Network Effects. The Journal

of Economic Perspectives 8(2): 93-115.

King J and West J (2002) Ma Bell’s orphan: US cellular telephony, 1947-1996,

Telecommunications Policy 26(3-4): 189-203.

Klepper S (1996) Entry, exit, growth, and innovation over the product life cycle,

American Economic Review 86: 562-583.

Langlois R (1993) External Economics and Economic Progress: the case of the

Microcomputer Industry. Business History 66: 1- 50.

Lehenkari J and Miettinen R (2002) Standardization in the construction of a large

technological system - the case of the Nordic mobile telephone system.

Telecommunications Policy 26(3-4): 109-127.

Leonard-Barton D (1992) Core capabilities and core rigidities: A paradox in managing

new product development. Strategic Management Journal (Special Issue) 13:

111-125.

Lewin A, Long C, Carroll T (1999) The coevolution of new organizational forms.

Organization Science 10(5): 535-550.

Lyytinen K and Fomin, W (2002) Achieving High Momentum in the Evolution of

Wireless Infrastructures: The Battle over the 1G solutions, Telecommunications

Policy 26: 149-190.

Meurling J and Jeans R (1997) The Ugly Duckling: Mobile phone from Ericsson –

putting people on speaking terms, Stockholm: Ericsson Mobile Communications AB.

Mintzberg H (1978) Structuring of Organizations. NY: Prentice Hall.

Murmann P (2003) Knowledge and Competitive Advantage: The Coevolution of Firms,

29

Technology, and National Institutions, Cambridge University Press.

Nadler D and Tushman M (1997) Competing by Design. Oxford University Press

Natsuno T (2003) i-mode Strategy. NY: John Wiley & Sons.

Nelson R (1995) Co-evolution of industry structure, technology, and supporting

institutions, and the making of comparative advantage. International Journal of the

Economics of Business 2(2): 171-184.

Nelson R (2002) Bringing institutions into evolutionary growth theory. Journal of

Evolutionary Economics 12: 17-28.

Nelson R and Winter S (2002) Evolutionary Theory of Economic Change. Boston:

Belknap.

Pelikan P (2003) Bringing institutions into evolutionary economics: another view with

links to changes in physical and social technologies. Journal of Evolutionary

Economics 13: 237-258.

Penrose E (1959) The theory of the growth of the firm. Oxford: Blackwell.

Reinhardt, A. and Ihlwan, M. (2005) Will Rewiring Nokia Spark Growth? Business

Week, February 14, pp. 20-22.

Rosenberg N (1976) Perspectives on Technology. Cambridge: Cambridge University

Press.

Scanlan M (2001) Hiccups in US spectrum auctions. Telecommunications Policy

25(10-11): 689-701.

Schumpeter J (1942) Capitalism, Socialism, and Democracy, NY: Harper & Row.

Shapiro C and Varian H (1999) Information Rules. Boston: Harvard Business School

Press.

Sigurdson J (2001) WAP OFF – Origin, Failure, and Future, Prepared for the

30

Japanese-European Technology Studies, University of Edinburgh, October 9, 2001.

Steinmueller E (2003) The Role of Technical Standards in Coordinating the Division of

Labour in Complex System Industries, in Prencipe A, Davies A, Hobday M (eds) The

Business of Systems Integration, NY: Oxford University Press.

Talley D (1982) Basic electronic switching for telephone systems. Hasbrouck Heights,

NJ: Hayden Book Co.

Teece D, Pisano G, Shuen A (2000) Dynamic Capabilities and Strategic Management,”

Strategic Management Journal 18: 509-534.

Utterback, J (1994) Mastering the dynamics of innovation: How companies can seize

opportunities in the face of technological change, Boston, MA: Harvard Business

School Press.

WSJ (Wall Street Journal) (1998) Qualcomm Clashes With Japan, Europe Over Cellular

Standard, November 17.

Ziman J (ed) (2000) Technological Innovations as an Evolutionary Process, Cambridge:

Cambridge University Press.

31

Table 1. Application of Durham’s (1991) Categories to the Mobile Phone Industry General Category In Mobile Phone Industry Unit of transmission Method of standard setting Sources of variation Different countries and different histories Mechanisms of transmission

Government agencies, multi-national firms, and standard setting institutions

Processes of transformation

Governments and firms adopt new routines in their standard setting processes to promote domestic growth and equipment suppliers

Sources of isolation Language, culture, geography, firm and government policies

Source: author’s analysis

32

Table 2. Evolution of Technology, Institutions, and Industry Structure (Leading Forms)

in Telecommunications Institutions Fixed

versus Mobile

Dates Generation/ Era

Method of Problem Solving Standard

setting Competition Policy

Industry Structure: Development Spending by

Until early 1980s

Regulated

Integral problem solving

Quasi vertical integration

Monopoly Service Provider

Fixed

After mid 1980s

Deregulated Modular problem Solving

Open standard setting at national level

Multiple licenses

Manufacturer

1970s and 1980s

First generation: analog

“ “ Some multiple licenses

Manufacturer

1980s and early 1990s

Second generation: digital

“ Open standard setting at trans-national level

Multiple licenses with some auctions

Manufacturer

Late 1990s

Third generation

“ Open standard setting at global level

Multiple licenses with more auctions

Manufacturer

Mobile

2000s Mobile Internet

Integral problem solving

Quasi vertical integration

No effect Service Provider?

Source: Adapted from (Brock, 1981; Fransman, 2002; Funk, 2002; Lyytinen and Fomin, 2002)

33

Table 3. Technological Change and Evolution of Standard Setting Approaches Technological

Generation Determination of Interface

Between Network and Phone Determination of Phone

Specifications Fixed Wireline Service providers determined everything until early 1980s

First generation mobile (analog)

Open process introduced in US and Scandinavia

Phone manufacturers in US and Scandinavia

Second generation mobile (digital)

Open process expanded to all parts of Europe and many parts of

Asia

Phone manufacturers in most of the world (except

Japan and Korea) Third generation mobile

Open process expanded to most parts of the world and dominated

by manufacturers

“

Mobile Internet In transition: failure of WAP Forum caused Western service providers to copy Japanese and Korean methods of determining

phone specifications and other interface standards Source: Adapted from (Brock, 1981; Fransman, 2002; Funk, 2002; Lyytinen and Fomin, 2002)

Table 4. Most Widely Used Air-Interface Standards

Generation of Technology

Communication Standard

Origin of Standard

Year in which Services were first started

1st

Generation Analog Cellular

NMT (Nordic Mobile Telephone) AMPS (Advanced Mobile Phone System)

Scandinavia North America

1981 1983

2nd

Generation Digital Cellular

GSM (Global System Mobile) TDMA (Time Division Multiple Access)PDC (Personal Digital Cellular) cdma (Code Division Multiple Access)

One

Europe U.S. Japan Qualcomm

1992 1992 1993 1994

3rd Generation Digital Cellular

Wide-Band CDMA cdma2000

Global Global

2001 2001

Sources: (Garrard, 1998; Funk and Methe, 2001; Funk, 2002; Lyytinen and 2002)

34

Figure 1. Evolution of Standard Setting Methods in Major Countries (Standards are in parentheses and arrows show evolution both within and

between generations)

Methods of Standard Setting

Generation Competition Open Process Quasi-of Between to Select Vertical

Technology Standards a single Standard Integration

Analog US (AMPS) Germany (C-Netz)Scandinavia France (RC-2000)

(NMT) Italy (RTMS)Japan (NTT)

Digital US (TDMA, Europe Japan (PDC) cdmaOne, (GSM) GSM)

Japan (PDC, cdmaOne)

Third US and Japan Europe Generation (cdma2000 (W-CDMA)

and W-CDMA)

Sources: (Funk and Methe, 2001; Funk, 2002; Lyytinen and Fomin, 2002; and author’s analysis)Abbreviations: NMT (Nordic Mobile Telephone); AMPS (Advanced Mobile Phone System);

NTT: Nippon Telephone and Telegraph; GSM: Global System Mobile; TDMA (Time Division Multiple Access); PDC (Personal Digital Cellular); cdma (Code Division Multiple Access); W-CDMA: Wide-Band cdma

35

Figure 2. Positive Feedback Between Market Growth, Manufacturer Development, and Support for Openness, Competition, and Patent Protection

Moresupport for

openness and competition

More supportfor openness and patent protection

Support for open standards,competition,

patent protection

Growth in markets that use open standards and

have competition

Developmentmoves from

service providers to other firms

(e.g., manufacturers)